WO2004010348A2

WO2004010348A2 - Method for determining the number of available transport seats in a computerized reservation system

Info

Publication number: WO2004010348A2
Application number: PCT/FR2003/050009
Authority: WO
Inventors: Denis Arnaud; Jean-Michel Sauvage; Valérie VIALE
Original assignee: Amadeus S.A.S.
Priority date: 2002-07-22
Filing date: 2003-06-13
Publication date: 2004-01-29
Also published as: AU2003255700B2; AU2003255700A1; NZ537856A

Description

"Method for determining the number of transport places available in a computer reservation system".

The present invention relates to a method for determining the number of transport places available in a computer reservation system.

This process will particularly find its application for reservation systems related to air transport. Other applications, and in particular to rail transport, are not however excluded from the present invention.

In the area of air transportation, in particular, the companies noted that, when they offer airlines operated several times a day, passengers often adopt a very specific way of booking their tickets. For short journeys where a return on the day is made, passengers generally book on the last possible flight and, when they can arrive at the airport earlier, they request to be transferred on a flight preceding that of their booking.

The same problem occurs when an airline adds a flight because it cannot further increase the capacity of another flight. For example, if a line is already taken by an Airbus 400® or a Boeing 747®, and the reservations have reached maximum capacity, the company is obliged to add an airplane. As a result, we find ourselves in a situation in which two flights coexist on the same line with an interval of 5 minutes. In such a case, the airline company probably does not want one of the flights to be full while the other has availability for passengers.

The computer reservation systems currently offered to airlines do not effectively prevent the behavior of certain passengers and therefore adapt to this type of behavior in order to offer the best transport services to users while ensuring financial returns. maximum for flights taking into account all the factors relating to it, in particular with regard to overbooking. The present invention makes it possible to remedy the drawbacks of the methods known up to now and proposes, to do this, a new method making it possible to calculate the availability capacities of airplanes (or other transport vectors) taking into account the existing availabilities on other flights. The invention therefore has the advantage of allowing airlines to have the possibility of offering a seat on classes of service (classes of transport services of two different flights) at the same time. In particular, the inventory system itself can calculate, according to the invention, the availability of a flight by including the availability of another flight and ensure that there is an interdependence of the reservations which are received on both thefts. If a seat is reserved on one of the flights, the availability of both flights will be updated, which ensures that the functionality offered here can be used in conjunction with the standard overbooking mechanisms.

In the case where a new aircraft is chartered because of the capacity limits of another flight, the present invention makes it possible to adjust the inventory system automatically to update the availability on the two flights when a reservation is made on one of them. In doing so, we always offer the possibility for the user to make a reservation on one of the flights if there is availability on the second. Of course, this system may require an awareness of the passengers to whom it will be proposed to make this transfer between two flights. However, if the two flights are particularly close in time, the transport service will not be degraded by the implementation of the functionality of the invention. The determination of availability thus proposed is operable dynamically, that is to say at each request addressed to the inventory databases on a request from a customer.

Other objects and advantages will appear during the following description of three preferred embodiments of the invention which are not, however, limiting.

The present invention relates to a method for determining the number of transport places available in a computer reservation system comprising means for storing data relating to transport services between two places and their current reservation status by class of service, method in which a number of places available locally for a given class of service of a transport service is determined at a predefined expected income level. According to the invention for a given class of service of a given transport service, the following steps are carried out:

- selection of at least one other class of service from a transport service;

- determination of the number of places available locally for the class of service of the transport service, at the predefined expected level of income;

- determination of a global number of places available for the given service class of the transport service given at the predefined expected level of income based on the different numbers of places available locally. Preferred variants of the method of the invention are introduced below: a single other class of service which belongs to another transport service is selected and the overall number of places available is determined by summing the numbers of places available locally of the two classes of service. each class of service is assigned a carry-over limit value corresponding to the maximum number of reservation requests on the class of service which can be carried over to places in other classes of service;

- an equal number of transferable booking requests are determined for each class of service: either zero if the number of places available locally for said class of service is positive, or opposite the number of places available locally for said class of service if said number is negative and its opposite is less than the limit value of carryover, or the carryover limit value if the number of places available locally is negative and its opposite is greater than or equal to the carryover limit value.

- an equal number of eligible reservation requests is determined for each service class: either zero if the number of places available locally for said service class is less than or equal to zero, or the number of places available locally for said class of service, each class of service is assigned an admission limit value corresponding to the maximum number of places in that class of service which can be used to transfer requests for reservations to other classes of service and the number of requests is assigned of eligible reservations an upper limit equal to the admission limit value. only one other class of service is selected which belongs to another transport service,

- the total admission capacity on the other service class for the given service class is determined by selecting the minimum value from the carry-over limit value for the given service class and the number of admissible reservation requests on said other class of service,

- the total carrying capacity from said other service class to the given service class is determined by selecting the minimum value from the number of reservation requests that can be carried over to the other service class and the number of admissible reservation requests on the given class of service,

- the overall number of places available is calculated: by adding the number of places available locally and the total admission capacity by subtracting the total carry-over capacity there, a transport service chain is formed having successive departure times and each comprising a selected service class,

- an index i is assigned to each transport service, the value of which increases with the time of departure,

- for each service class of a given transport service, the transport service service classes having a lower index towards which the requests for reservations on the service class of the given transport service can be transferred are selected. the total admission capacity for the service class is determined by selecting the minimum value from the carry-over limit value for the given service class and the sum of the numbers of eligible reservation requests for the transport service classes to which the given class of service is transferable. the total carrying capacity is determined from all the other service classes to a service class from the update of the number of eligible reservations to said service class, the overall number of places available is calculated: by adding the number of places available locally and the total capacity of admission by subtracting from it the total capacity of postponement, the steps of the process are carried out for each request for availability of a client.

The attached drawings are given by way of examples and are not limitative of the invention. They represent only one embodiment of the invention and will make it easy to understand it.

FIGS. 1 to 3 illustrate a first embodiment of the invention in which the possibility of transferring reservations for a late flight to previous flights is offered.

Figures 4 and 5 illustrate two other embodiments of the invention in which an interdependence is established between the reservations made on two flights. For the remainder of the description, the term “transport service” means any individual element making it possible, in a computer reservation system, to transport one or more users from a place of origin to a place of destination. In the particular case of application to air transport, these will be flights.

In addition, a flight may itself consist of several sections, in particular due to stopovers or connections. In the latter case, the transport service will correspond to a transport section.

It will also be noted that the present invention can also be applied for the transfer of reservations from a class of service (for example a business class of service) to another class of service (for example an economic class of service) of the same transport service.

We mean, moreover, for the remainder of the description, by the term local availability, the availability present at a given instant on a class of service of a given transport service without taking account of the reservation status of other transport services (therefore without having carried out the method of determining the invention). On the other hand, the global number of places available means the availability available on a service class of a given transport service at a given time, taking account of the reservation states on one or more other service classes. one or more other transport services (this is therefore the residual capacity on the given service class of the given transport service after implementation of the determination method according to the invention).

In the following description, three successive embodiments are described in the context of an application to air transport. In this context, the transport service is a flight segment and the service class is made up of the level of cabin service class taken by the user.

(e.g. economy service class or business service class).

The term expected income level is also understood to mean a return value of the ticket obtained by an expected profitability calculation. The availability calculation is understood to depend on the yield that one wishes to obtain from the ticket.

The method of the invention allows the realization of several operating scenarios. In a first case, the invention operates by advance booking.

In this context, it is applicable to any number of flights which travel on the same air transport line at different times throughout the day. This functionality makes it possible to take into account the behavior of users consisting in booking on the latest flights and then requesting a transfer.

According to another possibility of implementing the invention, a reservation twinning is carried out between two classes of services of two different flights. In this context, a single class of service is twinned between two flights, which allows access to the availability of one or the other of the flights without distinction.

Finally, we can also share a reservation which is the case where the two flights are most closely linked in terms of availability.

In this context, any of the classes of service of one of the flights can access the availabilities present on the same class of service of another flight.

First embodiment: advance booking.

FIG. 1 shows a case of this type of reservation with an implementation of four flights numbered 1, 2, 3 and 4 respectively which each include a specific class of service used here for the calculation of availability. It is the economic service class (M) which has been chosen for information only.

As shown, the four flights follow one another at different times of the day. The various arrows shown indicate possibilities for advancing the reservation, which constitute late flight reservation postponements to previous flights.

Each flight is assigned an index (i) which increases with the departure time of the flight. In the case of the illustrated example, it varies from 1 to 4. In addition, a parameter can be assigned to each flight indicating towards how many service classes of previous flights reservations carried over to the corresponding service class of the flight considered.

For the remainder of the description, this parameter is referred to as the advancement range. With reference to FIG. 1, in the case of flight 3, the advancement range is equal to 2, which means that the reservations for economy class M of flight 3 can be carried over to the identical service classes M for flights 1 and 2. Equivalently, reservations for service class M of the flight

4 can be carried over to the corresponding service classes of flights 2 and 3, taking into account that flight 4 also has a range of advancement equal to 2. Each flight, in the flight chain thus created, also has an additional parameter called carry-over limit hereinafter denoted sP ^max This value indicates the limit of the total number of passengers who can be carried over, here in advance, from the flight section considered on the previous flight sections.

In addition, each flight has another parameter called the admission limit value (hereinafter referred to as SA ^max ). The value of this parameter indicates the upper limit of the number of seats in a service class of a given flight which can accommodate carryovers from other flights, here later.

In the case of the example proposed here, only one class of service (here class of economic service) for each flight has the function of carry forward. The identity of the service class that has this functionality is clearly indicated by a field in the computer reservation system database.

Other parameters can also be taken into account, in particular the lower limit of expected income up to which a reservation deferral can be offered. This parameter makes it possible in particular to restrict access to the forward carry-forward functionality to certain users benefiting from high reservation levels. If this parameter is not specified, it is considered that any passenger is authorized to use the postponement functionality.

These various parameters and others which will be presented below are under the control of the manager of the computer reservation system via a graphical user interface if necessary.

We describe below how an overall availability for a given class of service of a given flight segment is calculated at a given level of expected income within the framework of a carry forward functionality. First, the local availability (AV _k ) of each class of service considered on each flight in the chain is conventionally calculated.

The number of transferable reservation requests (SP _Fik (Y)) and the number of admissible reservation requests (SA _F ik (Y)) is then determined for the service class considered.

The number of booking requests which can be carried over can be defined by the number of booking requests made on the given service class k of service which exceeds the physical capacity of said service class k and which can be carried over to other flights. Conversely, the number of admissible reservation requests represents the number of seats still physically available on a class of service of a flight and which can be used to receive transferable reservation requests from other flights.

It is easy to understand that if a given class of service has a non-zero number of transferable reservation requests, this means that the capacity of the service class considered is exceeded and, consequently, the number of reservation requests allowable on this class of service is zero. The converse is also true.

The definition of these parameters SP and SA will be well understood in view of FIG. 2.

Upper limits are also defined for the number of transferable reservation requests (SP) and the number of admissible reservation requests (SA). In this context, the admission limit (SA ^max ) corresponds to the maximum number of places in a given service class that can be used to carry over reservation requests made to other service classes (service classes of subsequent flights in this case). The carry-over limit parameter (SP ^max ) corresponds to the maximum number of reservation requests for a given class of service which can be carried over to service class seats on previous flights. An example is given below of the number of transferable and eligible reservation requests for an expected level of income Y given on a flight of index ι given with a given class of service _s :

SP _is (Y) = Min {SPis ^max , Max {0, - AV _is (Y *)}}

In which

Y * = Max {Y, expected income-minimum-de-reportj _S }

In this context, the minimum expected deferral income j _S is the minimum expected income for which it is possible to carry over from the class of service of the flight in question. This ensures that only the expected income above a certain level can benefit from the postponement functionality from the flight considered on previous flights.

Note that the number of eligible reservation requests (SA _JS (Y)) is an initial value. A parameter EFIj is then calculated for the service class of each flight in the chain. This makes it possible to determine the flight index of the chain which has the earliest departure time to which reservation requests on the service class of the flight i in question can be carried forward. The EFIj parameter is linked to the advancement range previously presented and visible in Figure 1.

If the advancement range of the flight i considered is greater than zero and that j is greater than 1, EF equals max (1, j - advancement range of i). Otherwise, EFIj equals zero.

It is thus possible to create an ordered list of the various flights making up the chain according to the value of their parameter EFIj.

The table below gives an example of how this table is made up:

VoM Vol 2 Vol 3 Vol 4 Vol 5 Vol 6 Vol 7 Vol 8

Theft index 1 2 3 4 5 6 7 8

Progress range 0 0 0 0 3 1 2 7

EFI 0 0 0 0 2 5 5 1

The flight order according to the EFI parameter would be here: flight 7, flight 6, flight 5 and flight 8.

The number of admissible reservation requests (SA) of the preceding flights is used to operate the allocation of the carry-over reservation requests (SP) of the subsequent flights so as to optimize the possibilities of postponement for a given flight. To do this, a calculation step is used, an example of which is given below with reference to FIG. 3. We start from the assumption that the advancement range of each flight here represented by a gray circle is equal to 3. The flight considered for the calculation is indicated by the most gray circle appearing substantially in the middle of the figure. The flights on the left are previous flights and the flights on the right are subsequent flights.

As shown in step 1 in FIG. 3, we first seek to allocate transferable reservation requests for flights preceding the flight in question on other flights, within the limit of the number of admissible reservation requests on flights on which are carried over. The arrows shown illustrate this allocation of reservations. It is thus possible to update and refine the values of the number of transferable (SP) and admissible (SA) reservation requests for these different flights.

We then consider flights with departure dates later than the flight in question. This is the step shown 2 in FIG. 3. We begin by seeking to allocate the booking requests that can be carried over from each of these different flights subsequent to the flight considered on previous flights but which have a greater or equal departure instant. to that of the flight in question.

We then seek to allocate the remaining transferable reservation requests on the other flights which have an earlier departure date. At this stage, we carried out all the possible postponements of the flights framing the flight in question so as to update their parameters SA and SP.

At this stage, allocation of transferable booking requests has not yet been made for the flight in question. However, at the end of this calculation, the values of the numbers of reservation requests admissible for the class of service considered on the other flights are known to allow the carry-over of the reservation requests which can be carried over from the flight considered.

We can then calculate the total admission capacity (ASD _ns (Y)) for the service class _s at the expected income given Y for the flight considered _n from the flights having departure dates earlier than the flight considered _n and to which reservation requests can be transferred. The following calculation can be made for this purpose:

TSAns (Y) = Min {SPnsmax, Σ "SA _ks (Y)} We can then calculate the total carry-over capacity (TSP _ns (Y)) which is the total number of reservation requests from later flights which can be carried over to the flight _n considered.

This total capacity is given by updating the number of admissible reservation requests (SA) for the flight considered during the carry-over allocation step previously described with reference to FIG. 3.

Indeed, the TSP capacity _ns (Y) is equal to the initial value of the number of eligible reservation requests (SA _ns ⁰ (Y)) minus the updated value of this number of eligible reservation requests (SA _ns (Y )) - The TSA and TSP values finally allow the calculation of the overall number of places available XFAV _ns (Y) for the flight _n considered.

This number is given by the following expression:

XFAV _ns (Y) = AV _ns (Y) + TSA _ns (Y) - TSP _nè (Y)

Second embodiment: twinning of flights.

According to this embodiment, we consider identical service classes k, k 'on two flights only.

This particular embodiment is illustrated in FIG. 4 in which the possibilities of reciprocal transfer between the economic service class M of flight 1 and the economic service class M of flight 2 have been represented by arrows.

As in the case of the first embodiment, the parameter SP ^{max is used} and possibly parameters linked to the limit expected income level authorizing the reservation carry-over functionality. The SA ^max parameter is not used because it is redundant with the SP ^max parameter of the paired service class.

An example is given below of determining the overall number of places available for a flight (here flight 1 F1) for the economic service class M at a predefined expected level of income Y. In this context, we start by identifying flight F2 to which flight F1 is paired. This can be done by a special field defining the identity of the paired flight for flight F1.

The local availabilities are then calculated for the service classes considered M of the two flights F1, F2. These local availabilities are hereinafter denoted av _f ι _k (Y) and av _f2k (Y).

This calculation step is similar to that performed in the first embodiment. In particular, standard calculation methods for determining local availability are used. We can then calculate the parameters for the number of transferable reservation requests and the number of admissible reservation requests (SA and SP) for the service classes M of the flights considered.

For each of the two flights F1, F2, the calculation is as follows:

SPi _s (Y) = Min {SP _is ^max , Max {0, - av _is (Y *)}} SA _is (Y) = Min {0, Max {0, av _is (Y)}}

In which

Y * = Max {Y, expected income-minimum-de-reportj _S }

It is then possible to calculate the total admission capacity TSAι _s (Y) on the twin flight F2 which is available for reports of the service class _s of the flight considered F1. This capacity is given by:

TSA _1s (Y) = Min {SPis ^max , SA _2s (Y)}

It is then possible to calculate the total carry-over capacity on the flight considered F1 from the service class _s of the paired flight F2. This capacity is given by: TSPι _s (Y) = Min {SP ₂ s, (Y), SA _1s (Y)}

We can finally determine the overall number of places available XFAV-is (Y) for flight F1 for the class of service _s considered at the predefined level of expected income Y.

This global number is given by the following formula: XFAVι _s (Y) = AViβ (Y) + TSA _1s (Y) - TSP _1s (Y)

Third embodiment: sharing of reservations.

Another embodiment of the invention is given below in the case of shared flights and for which the postponement of reservations is not limited to a given class of service on flights.

The number of transferable and admissible reservation requests is not constrained here by limit values SA ^max , SP ^max .

As in the case of the second embodiment, only two flights are considered. The principle of cooperation of the two flights is shown in Figure 5.

We start by calculating the number of places available locally for each of the two flights F1 and F2 for a class of service k considered. The overall number of places available for the service class k considered at the predefined expected level of expected income Y is then given by the sum of local availabilities according to the formula:

XFAV _1k (Y) = av _ik (Y) + av _2k (Y)

Claims

1. Method for determining the number of transport places available in a computer reservation system comprising means for storing data relating to transport services between two places and their current reservation status by class of service, method in which: determines, at a predefined expected income level (Y), a number of places available locally av _F ik (Y) for a given service class (k) of a given transport service (Fi) characterized in that for a given service class (k) of a given transport service (Fi), the following steps are carried out:

- selection of at least one other service class (k ') of a transport service (F _j );

- determination of the number of places available locally before _Fjk '(Y) for the service class (k') of the transport service (F _j ), at the predefined level of expected income (Y);

- determination of a global number of places available XFAV _Fik (Y) for the given service class (k) of the transport service (Fi) given at the predefined level of expected income (Y) according to the different numbers of places available locally (av _Fik (Y), av _Fj

2. Method according to claim 1 characterized in that

- select only one other class of service (k ') which belongs to another transport service (Fj) - determine the overall number of places available XFAV _Fik (Y) by summing the number of places available locally (av _Fik (Y), av _{Fjk '} (Y)) of the two service classes (k, k').

3. Method according to claim 1 characterized in that

- that each class of service is assigned a carry-over limit value (SP ^max ) corresponding to the maximum number of reservation requests for the class of service which can be carried over to places in other classes of service;

- that, for each class of service, an equal number of transferable reservation requests (SP (Y)) is determined: either zero if the number of places available locally for said class of service (k) is positive, or the opposite of the number of places available locally for said class of service (k) if said number is negative and its opposite is less to the carry-over limit value (SP ^ax ), or to the carry-over limit value (SP ^max ) if the number of places available locally is negative and its opposite is greater than or equal to the said carry-over limit value (SP ^max) ).

- that a number of eligible reservation requests (SA (Y)) is determined for each class of service: equal to zero if the number of places available locally for said class of service av (Y) is less than or equal zero, or the number of places available locally for said service class av k (Y) if this number is positive.

4. Method according to claim 3 characterized in that

- that an admission limit value (SA ^max ) corresponding to the maximum number of seats in the said service class which can be used to transfer reservations requests to other service classes is assigned to each service class, - that an upper limit equal to the admission limit value (SA ^max ) be assigned to the number of admissible reservation requests.

5. Method according to claim 3 characterized in that

- select only one other class of service (k ') which belongs to another transport service (F _j ), - determine the total admission capacity (TSA _k ) from the other class of service (k ') for the given service class (k) by selecting the minimum value from the carry-over limit value (SP ik ^max ) for the given service class (k) and the number of admissible reservation requests (SAFjk' ( Y)) on said other class of service (k '), - that the total carry-over capacity (TSP _k ) on said other class of service (k') is determined on the given class of service (k) by selecting the minimum value from the number of reservation requests that can be carried over for the other service class (k ') (SP _Fjk ' (Y)) and the number of admissible reservation requests for the given service class (k) (SA _Fik ( Y)), - that calculates the total number of places available XFAV _Fjk (Y) by adding the number of places available locally av _Fik (Y) and total capacity intake TSA _Fik (Y) by subtracting the total carry capacity TSP _Fik (Y)

6. Method according to claim 4 characterized in that

- forming a transport service chain (Fi) having successive departure times and each comprising a selected service class (k, k '),

- assign an index i to each transport service, the value of which increases with the time of departure,

- that for each service class (k) of a given transport service (Fi), the service classes (k ') of transport service having a lower index towards which the reservation requests on the class are selected of service of the given transport service (Fi) can be carried over.

7. Method according to claim 6 characterized in that the total admission capacity TSA _Fik (Y) for the service class (k) is determined by selecting the minimum value from the carry-over limit value (SP _Fik ^max ) of the given class of service (k) and the sum of the numbers of eligible reservation requests (SA _jk '(Y)) for the classes of service (k') of transport services (Fj) to which the class of service ( k) data is transferable.

8. Method according to claim 6 characterized in that the total TSP _Fik transfer capacity is determined from all the other service classes to a service class (k) from the updating of the number of requests eligible reservations to said service class (k).

9. Method according to claim 7 and claim 8 characterized in that

- calculate the overall number of places available XFAV _Fik (Y): by adding the number of places available locally before _F i _k (Y) and the total admission capacity TSAπ (Y) by subtracting the total capacity from postponement TSPπ _k (Y) 10. Method according to any one of claims 1 to 9, characterized in that that the steps of the process are carried out for each availability request from a client.